Machine Learning Basics (Stats & Data Distribution)

In the realm of machine learning, understanding the statistical properties of your data is crucial. This knowledge helps you make informed decisions about which algorithms to use and how to preprocess your data effectively. In this tutorial, we'll explore key concepts such as mean, median, mode, standard deviation, percentiles, normal distribution, scatter plots, and how to analyze data distributions using Python.

Introduction

Statistics provides the tools to understand and interpret data. Whether you're analyzing stock market trends or predicting customer behavior, a solid grasp of statistical measures is essential. In this tutorial, we'll dive into fundamental statistics concepts and learn how to apply them using Python libraries like NumPy and Matplotlib.

Core Concepts

Mean, Median, Mode

The mean, median, and mode are basic measures of central tendency that help you understand the center of your data distribution.

Mean

The mean is the average value of a dataset. It's calculated by summing all the values and dividing by the number of observations.

1import numpy as np
2 
3data = [1, 2, 3, 4, 5]
4mean_value = np.mean(data)
5print("Mean:", mean_value)

Mode

The mode is the value that appears most frequently in a dataset.

1from scipy import stats
2 
3data = [1, 2, 2, 3, 4]
4mode_value = stats.mode(data)
5print("Mode:", mode_value.mode[0])

Normal Distribution

A normal distribution, also known as a Gaussian distribution, is symmetric and bell-shaped. Many natural phenomena follow this distribution.

1import numpy as np
2import matplotlib.pyplot as plt
3 
4data = np.random.normal(loc=0, scale=1, size=1000)
5plt.hist(data, bins=30, density=True, alpha=0.6, color='g')
6plt.title('Normal Distribution')
7plt.xlabel('Value')
8plt.ylabel('Density')
9plt.show()

Scatter Plots

Scatter plots are used to visualize the relationship between two variables. Each point on the plot represents an observation.

1import numpy as np
2import matplotlib.pyplot as plt
3 
4x = np.random.rand(50)
5y = 2 * x + np.random.normal(loc=0, scale=0.1, size=50)
6 
7plt.scatter(x, y)
8plt.title('Scatter Plot')
9plt.xlabel('X')
10plt.ylabel('Y')
11plt.show()

Practical Example

Let's analyze a dataset to understand its distribution and relationships.

1import numpy as np
2import matplotlib.pyplot as plt
3from scipy import stats
4 
5# Generate sample data
6np.random.seed(0)
7data = np.random.normal(loc=50, scale=10, size=1000)
8 
9# Calculate statistics
10mean_value = np.mean(data)
11median_value = np.median(data)
12mode_value = stats.mode(data).mode[0]
13std_dev = np.std(data)
14percentile_25 = np.percentile(data, 25)
15 
16print(f"Mean: {mean_value}")
17print(f"Median: {median_value}")
18print(f"Mode: {mode_value}")
19print(f"Standard Deviation: {std_dev}")
20print(f"25th Percentile: {percentile_25}")
21 
22# Plot histogram
23plt.hist(data, bins=30, density=True, alpha=0.6, color='g')
24plt.title('Data Distribution')
25plt.xlabel('Value')
26plt.ylabel('Density')
27plt.show()

Summary

What's Next?

In the next tutorial, we'll explore linear and polynomial regression, which are fundamental techniques for modeling relationships between variables. Understanding these concepts will help you build predictive models that can make accurate predictions based on data.

Stay tuned!